Based on newly developed algorithms, the article presents results of simulation modeling of main indicators of productivity, reliability and operation costs of a storage system for deduplicated data on solid-state media. The influence of input control parameters of the system on the output performance indicators is analyzed, which allows proposing optimal parameters and structure of a data storage system during its design. As a result of the study, a simulation system was developed and implemented using R and Python programming languages, based on the proposed simulation model and algorithms. The simulation model of a queuing system is implemented as a multi-server model for processing requests with unlimited queues. It allows simulating the processes of reading and writing information in a data storage system and evaluating performance indicators, such as data transfer speed in read and write modes, as well as the system response time. A distinctive feature of the simulation model is a newly developed algorithm for calculating the estimated requests processing time by the deduplication system, which allows for the first time to quantify the effect of applying this technology. The simulation results confirmed the effectiveness of using the deduplication technology for storage performance of a data storage system. To solve the inverse design problem, an algorithm was developed for finding the most effective storage settings. The proposed algorithm allows finding solutions that are optimal in terms of a combination of factors — reliability, speed and cost of data storage. The results obtained during simulation modeling correspond to results of practice experiments, which confirms the adequacy of the model. The developed toolkit allows designing a data storage system of required volume and storage duration with minimum storage costs with given performance and reliability indicators.